AI detection tools invite a scan-and-shame ritual. A score appears: human, AI-assisted, or AI. Readers treat it as a ruling on the value of the writing. The tool itself makes no such claim. It examines patterns of syntax and word choice, not whether the words are worth reading.

This is an old error in a new form. Process-focused attacks were once aimed at Lincoln’s Gettysburg Address and at the sources of Martin Luther King Jr.’s most famous speech. The tallies changed nothing. The words survived. The same arithmetic is now being applied to newsletters and essays.

Scientific invention requires more than pattern recognition or formal proof. Following Einstein’s own account of discovery, it depends on an intuitive jump from experience to new axioms. Only after that jump does deduction and verification become possible.

Large language models have become strong at induction (extracting patterns from data) and are rapidly improving at deduction (reasoning from given premises). What they still lack is the capacity for the abductive jump itself: the generation of novel explanatory principles when existing ones are insufficient.

Highly intelligent people frequently get trapped by their own elegant mental models. They fall in love with beautiful theories and frameworks, prioritising intellectual coherence and elegance over messy real-world evidence and results.

This pattern repeats across domains. Behavioural scientists predictions on motivation performed worse than random chance. Trained therapists often do no better than casual conversations. Most hedge fund managers fail to beat the market. Even diversity training frequently increases bias instead of reducing it.